CEUS-Net: Lesion Segmentation in Dynamic Contrast-Enhanced Ultrasound with Feature-Reweighted Attention Mechanism

Wan, Pengfei; Chen, Fang; Zhu, Xiaojin; Liu, Chunrui; Zhang, Yidan; Kong, Wencheng; Zhang, Daoqiang

doi:10.1109/isbi45749.2020.9098655

Cited by 5 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors divided the algorithm in two categories: one is for the enhancement representation learning, and the other is for hierarchical lesion recognition. For enhancement learning, they are segmenting the frames using a CEUS-Net previously developed in [ 29 ]. The network consists of convolutional layers, local pooling layers and a final BatchNorm and ReLu activation function.…”

Section: State Of the Artmentioning

confidence: 99%

“…The training procedure will consider just 11,969 ROI pictures, coming from the rest of 90 patients, from the total of 12119. The trained model will predict the correct class label for the test set if at least 1/5th + 1 of the predictions are correct, e.g., [31 FNH,29 In Figures 9 and 10, respectively, the individual accuracies for the case of unbalanced number of examples per class of one of the five folds/experiments and, respectively, the average experiment accuracies, are presented. Hard voting scheme is employed for calculating the decision.…”

Section: Voting Schemementioning

confidence: 99%

See 1 more Smart Citation

Deep Neural Architectures for Contrast Enhanced Ultrasound (CEUS) Focal Liver Lesions Automated Diagnosis

Căleanu

Sîrbu

Simion

2021

Sensors

View full text Add to dashboard Cite

Computer vision, biomedical image processing and deep learning are related fields with a tremendous impact on the interpretation of medical images today. Among biomedical image sensing modalities, ultrasound (US) is one of the most widely used in practice, since it is noninvasive, accessible, and cheap. Its main drawback, compared to other imaging modalities, like computed tomography (CT) or magnetic resonance imaging (MRI), consists of the increased dependence on the human operator. One important step toward reducing this dependence is the implementation of a computer-aided diagnosis (CAD) system for US imaging. The aim of the paper is to examine the application of contrast enhanced ultrasound imaging (CEUS) to the problem of automated focal liver lesion (FLL) diagnosis using deep neural networks (DNN). Custom DNN designs are compared with state-of-the-art architectures, either pre-trained or trained from scratch. Our work improves on and broadens previous work in the field in several aspects, e.g., a novel leave-one-patient-out evaluation procedure, which further enabled us to formulate a hard-voting classification scheme. We show the effectiveness of our models, i.e., 88% accuracy reported against a higher number of liver lesion types: hepatocellular carcinomas (HCC), hypervascular metastases (HYPERM), hypovascular metastases (HYPOM), hemangiomas (HEM), and focal nodular hyperplasia (FNH).

show abstract

Section: State Of the Artmentioning

confidence: 99%

Section: Voting Schemementioning

confidence: 99%

Deep Neural Architectures for Contrast Enhanced Ultrasound (CEUS) Focal Liver Lesions Automated Diagnosis

Căleanu

Sîrbu

Simion

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Both conventional methods and deep learning techniques were also applied for performing tumor recognition and segmentation within contrast-enhanced medical images [ 34 , 35 , 36 , 37 ], respectively, within combinations of different medical image modalities [ 26 , 38 ], the CNN techniques having an important role in this context. The most relevant approaches in this field are described below.…”

Section: Introductionmentioning

confidence: 99%

“…In addition, some of the reports demonstrated that CEUS manifested a specific sensitivity to metastases smaller than 10 mm, improving the corresponding sensitivity by approximately 50%, compared with the classical US. A method that performed automatic tumor segmentation within a dynamic sequence of CEUS images was described in [ 36 ]. A new CNN architecture, named CEUS-Net, was defined and assessed.…”

Section: Introductionmentioning

confidence: 99%

Hepatocellular Carcinoma Automatic Diagnosis within CEUS and B-Mode Ultrasound Images Using Advanced Machine Learning Methods

Mitrea

Badea

Mitrea

et al. 2021

Sensors

View full text Add to dashboard Cite

Hepatocellular Carcinoma (HCC) is the most common malignant liver tumor, being present in 70% of liver cancer cases. It usually evolves on the top of the cirrhotic parenchyma. The most reliable method for HCC diagnosis is the needle biopsy, which is an invasive, dangerous method. In our research, specific techniques for non-invasive, computerized HCC diagnosis are developed, by exploiting the information from ultrasound images. In this work, the possibility of performing the automatic diagnosis of HCC within B-mode ultrasound and Contrast-Enhanced Ultrasound (CEUS) images, using advanced machine learning methods based on Convolutional Neural Networks (CNN), was assessed. The recognition performance was evaluated separately on B-mode ultrasound images and on CEUS images, respectively, as well as on combined B-mode ultrasound and CEUS images. For this purpose, we considered the possibility of combining the input images directly, performing feature level fusion, then providing the resulted data at the entrances of representative CNN classifiers. In addition, several multimodal combined classifiers were experimented, resulted by the fusion, at classifier, respectively, at the decision levels of two different branches based on the same CNN architecture, as well as on different CNN architectures. Various combination methods, and also the dimensionality reduction method of Kernel Principal Component Analysis (KPCA), were involved in this process. These results were compared with those obtained on the same dataset, when employing advanced texture analysis techniques in conjunction with conventional classification methods and also with equivalent state-of-the-art approaches. An accuracy above 97% was achieved when our new methodology was applied.

show abstract

CEUSegNet: A Cross-Modality Lesion Segmentation Network for Contrast-Enhanced Ultrasound

Meng

Zhu

Fan

et al. 2022

2022 IEEE 19th International Symposium on Biomedical Imaging (ISBI)

View full text Add to dashboard Cite

CEUS-Net: Lesion Segmentation in Dynamic Contrast-Enhanced Ultrasound with Feature-Reweighted Attention Mechanism

Cited by 5 publications

References 8 publications

Deep Neural Architectures for Contrast Enhanced Ultrasound (CEUS) Focal Liver Lesions Automated Diagnosis

Deep Neural Architectures for Contrast Enhanced Ultrasound (CEUS) Focal Liver Lesions Automated Diagnosis

Hepatocellular Carcinoma Automatic Diagnosis within CEUS and B-Mode Ultrasound Images Using Advanced Machine Learning Methods

CEUSegNet: A Cross-Modality Lesion Segmentation Network for Contrast-Enhanced Ultrasound

Contact Info

Product

Resources

About